Title: A linear regression based prediction model for load distribution and quality of service improvement with different resource utilisation in cloud environment

Authors: Gopa Mandal; Santanu Dam; Kousik Dasgupta; Paramartha Dutta

Addresses: Department of CSE, Jalpaiguri Govt. Engineering College, Jalpaiguri, 735102, India ' Department of HE, Netaji Subhas Open University, Salt Lake, Kolkata, 700064, India ' Department of CSE, Kalyani Govt. Engineering College, Kalyani, 741235, India ' Department of Computer and System Sciences, Visva- Bharati University, Shantiniketan, 731 235, India

Abstract: Cloud computing is a delivery-based consumption model that relies on the internet. The use of cloud enabled devices is increasing rapidly. So, to maintain quality of service (QoS), throughput of the entire system with service level agreements (SLAs) is a major concern between the service providers and the end users. Alternative techniques for virtual machine (VM) consolidation and proper workload allocation may be beneficial. This study proposes a linear regression-based prediction model for load distribution and QoS improvement. The model aims to enhance system throughput and QoS by predicting resource utilisation levels using historical consumption data. Experiments conducted using the CloudSim and CloudAnalyst platforms demonstrate positive results, outperforming existing methodologies. The study also evaluates service level agreement violation (SLAV) and delays to assess the QoS provided by the cloud service provider (CSP). Overall, this research contributes to the enhancement of QoS in cloud and cloud enabled systems like the Internet of Things and the Cloud of Things (CoT) and addresses the challenges of optimising resource utilisation while ensuring QoS.

Keywords: CoT; Cloud of Things; IoT; Internet of Things; VM consolidation; cloud computing; QoS; quality of service; CloudSim; CloudAnalyst; linear regression.

DOI: 10.1504/IJCNDS.2025.145920

International Journal of Communication Networks and Distributed Systems, 2025 Vol.31 No.3, pp.277 - 317

Received: 25 Mar 2024
Accepted: 19 May 2024
Published online: 30 Apr 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article